Frontier AI Safety Researcher

OpenAI; Greater London; On-site; GBP 218,000 - 328,000; Full time

Job summary

OpenAI is seeking exceptional researchers to derisk frontier models by developing novel safety mitigations and applying techniques from interpretability, control, and alignment. You will help define what a safe AI system should look like and contribute to OpenAI’s mission to build safe AGI.

This role requires strong technical depth, cross-functional collaboration with misalignment, cybersecurity, and biology experts to create a scalable and effective safety stack for deployed models.

Qualifications

• 2+ years of experience in the field of AI safety, especially in areas like RLHF, human-AI collaboration, interpretability, or control.
• Hold a Ph.D. or other degree in computer science, machine learning, or a related field.
• Thrive in environments involving large-scale AI systems.
• Bring 4+ years of research engineering experience and proficiency in Python or similar languages.

Responsibilities

• Identify emerging AI safety risks and develop methodologies to mitigate them.
• Build and refine evaluations to assess risks, collaborating with domain experts.
• Set research directions to improve safety, alignment, and robustness of AI systems.
• Develop best practices and guidelines for AI safety for OpenAI and the industry.
• Evaluate and design red-teaming pipelines to test end-to-end safety systems.

Job description

OpenAI is seeking exceptional researchers to derisk frontier models by developing novel safety mitigations and applying techniques from interpretability, control, and alignment. You will help define what a safe AI system should look like and contribute to OpenAI’s mission to build safe AGI.

This role requires strong technical depth, cross-functional collaboration with misalignment, cybersecurity, and biology experts to create a scalable and effective safety stack for deployed models.